Search

Mathematics and Music are closely connected and their multifaceted relationship has been explored since ancient times. Pythagoras was one of the first who discovered and formalized one of those connections by studying the arithmetic of musical intervals, expressing them by numerical ratios and relating ratios to consonance/dissonance notions, being the Pythagorean tuning and scale based on arithmetic principles. Another interesting link between Mathematics and Music is the geometric approach to musical composition. Geometric patterns are present in different musical style compositions and some composing techniques, such as the 12-tone technique, use geometric transformations, involving also applications of modular arithmetic and set theory to Music. The aim of this work is to present existing relations between Mathematics and Music and to propose automatic exercise generation for exploring and studying those relations. The exercises are generated using the system MVGEN and the LaTeX package MusiXTeX. The generation process and the automatically generated exercises can be used for creating learning and assessment materials for education in Mathematics and Music, linking Science and Art., (undefined)

Political debate - in its essence - carries a robust emotional charge, and social media have become a vast arena for voters to disseminate and discuss the ideas proposed by candidates. The Brazilian presidential elections of 2018 were marked by a high level of polarization, making the discussion of the candidates’ ideas an ideological battlefield, full of accusations and verbal aggression, creating an excellent source for sentiment analysis. In this paper, we analyze the emotions of the tweets posted about the presidential candidates of Brazil on Twitter, so that it was possible to identify the emotional profile of the adherents of each of the leading candidates, and thus to discern which emotions had the strongest effects upon the election results. Also, we created a model using sentiment analysis and machine learning, which predicted with a correlation of 0.90 the final result of the election., FCT – Fundação para a Ciência e Tecnologia within the Project Scope: UID/CEC/00319/2019

In this paper, we'll discuss a simple approach to integrating musical events, such as notes or chords, into a programming language. This means treating music sequences as a first class citizen. It will be possible to save those sequences into variables or play them right away, pass them into functions or apply operators on them (like transposing or repeating the sequence). Furthermore, instead of just allowing static sequences to be generated, we'll integrate a music keyboard system that easily allows the user to bind keys (or other kinds of events) to expressions. Finally, it is important to provide the user with multiple and extensible ways of outputing their music, such as synthesizing it into a file or directly into the speakers, or writing a MIDI or music sheet file. We'll structure this paper first with an analysis of the problem and its particular requirements. Then we will discuss the solution we developed to meet those requirements. Finally we'll analyze the result and discuss possible alternative routes we could've taken., (undefined)

This paper presents an unified approach for discovering genealogical information. It presents a frameworks for storing information concerning ancestors, locations, dates and documents. It also intends to provide a framework that is able to perform inference concerning dates by using constraints and for handling relations, locations and sources. The DSL presented also aims to help users store information from heterogeneous sources along with the evidence contained therein., FCT – Fundação para a Ciência e Tecnologia within theProject Scope: UID/CEC/00319/2013

O sistema Passarola é um sistema de geração automática de exercícios que foi desenvolvido na Universidade do Minho para apoiar o ensino, a aprendizagem e a auto-aprendizagem no ensino superior. O sistema Passarola permite a geração de exercícios de várias áreas de ensino, como por exemplo Matemática, Física, Música, Geografia, e com respostas que vão além dos tipos simples como escolha múltipla ou soluções verdadeiro/falso. É possível obter respostas com vários passos de transformação e etapas de resolução. Para a compreensão de certos conceitos, por exemplo da área da Matemática, é importante a geração de exercícios com gráficos dinâmicos. Estes podem ser construídos na Passarola com os programas TikZ, LaTeX e Maxima. Neste trabalho apresentam-se exemplos de exercícios de diferentes áreas da Matemática de nível universitário introdutório, que foram implementados no sistema Passarola e que podem ser utilizados para criar fichas de aprendizagem e auto-avaliação, testes e exames.

When dealing with chatbots, domain identification is an important feature to adapt the interactions between user and computer in order to increase the reliability of the communication and, consequently, the audience and decrease its rejection avoiding misunderstandings. In order to adapt to different domains, the writing style will be different for the same author. For example, the same person in the role of a student writes to his professor in a different style than he does for his brother. This article presents a process that uses sentiment analysis to identify the average emotional profile of the communication scenario where the conversation is done. Using Natural Language Processing and Machine Learning techniques, it was possible to obtain an index of 96.21% of correct classifications in the identification of where these communications have occurred only analysing the emotional profile of these texts., This work has been supported by COMPETE: POCI-01-0145-FEDER-0070 43and FCT – Fundação para a Ciência e Tecnologia within the Project Scope: UID/CEC/ 00319/2013

One of the major problems when using lexicon in sentiment analysis is that they do not cover all possible words in a text and frequently they miss the more expressive to describe the emotions of the text's author efficiently. This problem occurs because people in non-official, on formal channels, communicate using slangs, neologisms, new patterns based on abbreviations (as "aka", "brb" and "asap") and the different meanings, making challenging to analyse texts using a finite subset of a language. This is a problem because some unknown words can completely change the meaning of a sentence, producing misunderstandings. In this paper we present an approach to expand an emotional lexicon for a specific author, producing a customised lexicon which represents how the author "feels" the words. In our experiments, we got an increase of 35.34% and 107.02% in the dictionary size when compared to the original lexicon using two different authors, and identifying different emotions from the same text according to each author's lexicon, i.e. interpreting the text according to the author's "point of view"., This work has been supported by COMPETE: POCI-01-0145-FEDER-0070 43 and FCT - Fundação para a Ciência e Tecnologia within the Project Scope UID/CEC/00319/2013.

Writing style is considered the manner how an author expresses his thoughts, influenced by language characteristics of an individual, period, school, or nation. Most of the times, this writing style can identify the author. Yet, one of the most famous examples comes from 1914 in Portuguese literature, with Fernando Pessoa and his heteronyms Alberto Caeiro, Álvaro de Campos and Ricardo Reis, who had completely different writing styles and led people to believe that they were different individuals. So, the discussion about authorship identification already exists along a century. Currently, there are several alternatives to identify authors of text, however, these solutions do not consider the emotion contained in the text as source of information in the writing style. This paper is about a process to analyse the emotion contained in social media messages as Facebook (http://www.facebook.com) in order to identify the author’s emotional profile and use it to improve the ability to predict the authors of the messages. Using preprocessing techniques, lexicon-based approaches and machine learning, we achieved an authorship identification improvement around 5% in the whole dataset and more than 50% in specific authors, when considering the emotional profile on the writing style., This work has been supported by COMPETE: POCI-01-0145-FEDER-0070 43 and FCT – Fundação para a Ciencia e Tecnologia within the Project Scope UID/CEC/00319/2013.

It is well known the relevance of accompany chords but there is a lack of tools capable of automatically generating sound from them. In this paper we describe a domain specific language (Abcl) aimed to be a prototyping environment for new experimental music operators. Currently Abcl: (1) adds support for accompany chords (chordmode, instruments, chord-lines); (2) adds clearer support for percussion (drums, drum-machine) (3) adds a support for variables and functions. Abcl tool is a syntactic-preprocessor that produces Abc. The DSLToolkit, used to create Abcl, is also briefly presented and discussed in the paper., (undefined)

It is well known that (1) problems have a structure, (2) often this structure comes from its input, output or internal objects, (3) this structure organizes the problem resolution. Decomposition of a problem by its structure is a typical path in problem resolution. In this article we will do the inverse: composition of simple problems using structure-builder operators. We will start by (1) discuss problem structures, (2) select a set of important structure-builder operators, (3) study their properties, (4) apply them to a simple math case-study., Uminho - Universidade do Minho (PEst-OE/MAT/UI0013/2014), info:eu-repo/semantics/publishedVersion

In this paper we present a metagrammar based algorithm for exercise generation in the domain of context-free grammars. We also propose a probabilistic assessment algorithm based on a new identity theorem for formal series, a matrix version of the well-known identity theorem from the theory of analytic functions., (undefined), info:eu-repo/semantics/publishedVersion

This paper presents two simple experiments performed in order to enlarge the coverage of PULO, a Lexical Ontology, based and aligned with the Princeton WordNet. The first experiment explores the triangulation of the Galician, Catalan and Castillian wordnets, with translation dictionaries from the Apertium project. The second, explores Dicionário-Aberto entries, in order to extract synsets from its definitions. Although similar approaches were already applied for different languages, this document aims at documenting their results for the PULO case.

Definido um enredo concreto, há, frequentemente, uma grande quantidade e variedade de exercícios que sobre ele podemos escrever. Neste artigo, apresenta-se um algoritmo que permite automatizar este processo. Para gerar exercícios matemáticos, parte-se das relações entre as variáveis presentes no enredo, calcula-se os possíveis exercícios, analisa-se a complexidade do seu processo de resolução e a sua viabilidade. O algoritmo recorre a bases de Gröbner para determinar se o exercício é resolúvel e um possível caminho de resolução. Com base na análise dos resultados obtidos, torna-se possível criar uma base de exercícios ligada a esse enredo., Given a concrete scenario, there is frequently a huge quantity and variety of exercises that we can write about it. In this article, we present an algorithm which allows us to automate this process. To generate mathematical exercises, we start from the relations between the variables present in the scenario, we calculate the possible exercises, we analise the complexity of its resolution process and its viability. The algorithm uses Gröbner bases to determine if the exercise is solvable and a possible path of resolution. Based on the analysis of the obtained results, its possible to create an exercises base related to that scenario.

In the later years the amount of freely available multilingual corpora has grown in an exponential way. Unfortunately the way these corpora are made available is very diverse, ranging from simple text files or specific XML schemas to supposedly standard formats like the XML Corpus Encoding Initiative, the Text Encoding Initiative, or even the Translation Memory Exchange formats. In this document we defend the usage of Translation Memory Exchange documents, but we enrich its structure in order to support the annotation of the documents with different information like lemmas, multi-words or entities. To support the adoption of the proposed formats, we present a set of tools to manipulate the different formats in an agile way.

Software maintainers are often challenged with source code changes to improve software systems, or eliminate defects, in unfamiliar programs. To undertake these tasks a sufficient understanding of the system, or at least a small part of it, is required. One of the most time consuming tasks of this process is locating which parts of the code are responsible for some key functionality or feature. This paper introduces Conclave, an environment for software analysis, that enhances program comprehension activities. Programmers use natural languages to describe and discuss the problem domain, programming languages to write source code, and markup languages to have programs talking with other programs, and so this system has to cope with this heterogeneity of dialects, and provide tools in all these areas to effectively contribute to the understanding process. The source code, the problem domain, and the side effects of running the program are represented in the system using ontologies. A combination of tools (specialized in different kinds of languages) create mappings between the different domains. Conclave provides facilities for feature location, code search, and views of the software that ease the process of understanding the code, devising changes. The underlying feature location technique explores natural language terms used in programs (e.g. function and variable names); using textual analysis and a collection of Natural Language Processing techniques, computes synonymous sets of terms. These sets are used to score relatedness between program elements, and search queries or problem domain concepts, producing sorted ranks of program elements that address the search criteria, or concepts respectively. © Nuno Ramos Carvalho, José João Almeida, Maria João Varanda Pereira, and Pedro Rangel Henriques., (undefined), info:eu-repo/semantics/publishedVersion

Parallel text alignment is a key procedure in the automated translation area. A large number of aligners have been presented along the years, but these require that the target resources have been pre-prepared for alignment (either manually or automatically). It is rather normal to encounter mixed language documents, that is, documents where the same information is written in many languages (Ex: manuals of electronic devices, touristic information, PhD thesis with dual language abstracts, etc). In this article we present MLT-prealigner: a tool aimed at helping those that need to process mixed texts in order to feed alignment tools and other related language systems., (undefined), info:eu-repo/semantics/publishedVersion

Series : OASIcs - Open access series in informatics, ISSN 2190-6807, vol. 38, One of the first tasks when building a Natural Language application is the detection of the used language in order to adapt the system to that language. This task has been addressed several times. Nevertheless most of these attempts were performed a long time ago when the amount of computer data and the computational power were limited. In this article we analyze and explain the use of a neural network for language identification, where features can be extracted automatically, and therefore, easy to adapt to new languages. In our experiments we got some surprises, namely with the two Chinese variants, whose forced us for some language-dependent tweaking of the neural network. At the end, the network had a precision of 95%, only failing for the Portuguese language., (undefined)

Texts aligned alongside their translation, or Parallel Corpora, are a very widely used resource in Computational Linguistics. Processing these resources, however, is a very intensive, time consuming task, which makes it a suitable case study for High Performance Computing (HPC). HPC underwent several recent changes, with the evolution of Heterogeneous Platforms, where multiple devices with different architectures are able to share workload to increase performance. Several frameworks/toolkits have been under development, in various fields, to aid the programmer in extracting more performance from these platforms. Either by dynamically scheduling the workload across the available resources or by exploring the opportunities for parallelism. However, there is no toolkit targeted at Computational Linguistics, more specifically, Parallel Corpora processing. Parallel Corpora processing can be a very time consuming task, and the field could definitely use a toolkit which aids the programmer in achieving not only better performance, but also a convenient and expressive way of specifying tasks and their dependencies., (undefined), info:eu-repo/semantics/publishedVersion

Probabilistic Translation Dictionaries are around for some time, but there is a lack of a formal definition for their structure and base operations. In this article we start by discussing what these resources are, what researchers are using them for, and what tools can be used to create this them. Including a formal definition and a proposal for a XML schema for dictionaries interchange. Follows a discussion of a set of useful operations that can be performed over probabilistic translation dictionaries, like union, intersection, domain restriction and compo- sition. Together with this algebra formalization some insights on the operations usefulness and application are presented., This work is partially supported by Per-Fide. The Per-Fide project is supported in part by a grant (Reference No. PTDC/CLEL-LI/108948/2008) from the Portuguese Foundation for Science and Technology and it is co-funded by the European Regional Development Fund.

ABC is a simple, yet powerful, textual musical notation. This paper presents ABC::DT, a rule-based domain-specific language (Perl embedded), designed to simplify the creation of ABC processing tools. Inspired by the Unix philosophy, those tools intend to be simple and compositional in a Unix filters' way. From ABC::DT's rules we obtain an ABC processing tool whose main algorithm follows a traditional compiler architecture, thus consisting of three stages: 1) ABC parser (based on abcm2ps' parser), 2) ABC semantic transformation (associated with ABC attributes), 3) output generation (either a user defined or system provided ABC generator)., (undefined), info:eu-repo/semantics/publishedVersion

Perl is known for its versatile regular expressions. Nevertheless, using Perl regular expressions for creating fast lexical analyzer is not easy. As an alternative, the authors defend the automated generation of the lexical analyzer in a well known fast application (flex) based on a simple Perl definition in the syntactic analyzer. In this paper we extend the syntax used by Parse::Yapp, one of the most used parser generators for Perl, making the automatic generation of flex lexical scanners possible. We explain how this is performed and conclude with some benchmarks that show the relevance of the approach.

Concept location is a common task in program comprehension techniques, essential in many approaches used for software care and software evolution. An important goal of this process is to discover a mapping between source code and human oriented concepts. Although programs are written in a strict and formal language, natural language terms and sentences like identifiers (variables or functions names), constant strings or comments, can still be found embedded in programs. Using terminology concepts and natural language processing techniques these terms can be exploited to discover clues about which real world concepts source code is addressing. This work extends symbol tables build by compilers with ontology driven constructs, extends synonym sets defined by linguistics, with automatically created Probabilistic SynSets from software domain parallel corpora. And using a relational algebra, creates semantic bridges between program elements and human oriented concepts, to enhance concept location tasks.

Most existing programming languages can be categorized as general purpose programming languages, meaning that they can be used to implement solutions for any given domain. They are not, in any way, optimized for a specific set of problems. In contrast, Domain Specific Languages (DSL) are used to solve specific problems in a well defined domain. DSL are optimized to a particular set of problems, but they lack support for a wide range of operations that are required when dealing with real world problems. So, in a perfect world, we would like to implement applications using a general purpose programming language, but use a set of different DSL to handle specific domains' tasks. In this paper we describe a DSL named Ontology Manipulation Language (OML), designed to describe operations over with ontologies. Programs can be written using only the OML syntax and be executed independently. OML syntax was designed to deal with ontologies and the language itself is optimized to perform these tasks, which means that other relatively simpler tasks can not be easily done. To overcome this challenge a mechanism was developed so that you can weave small snippets of OML code inside Perl programs, meaning we have the power of OML to manipulate ontologies and, at the same time, all the paraphernalia of modules that Perl offers to handle everything else.

The eXtensible Mark-up Language (XML) is probably one of the most popular markup languages available today. It is very typical to find all kind of services or programs representing data in this format. This situation is even more common in web development environments or Service Oriented Architectures (SOA), where data flows from one service to another, being consumed and produced by an heterogeneous set of applications, which sole requirement is to understand XML. This workflow of data represented in XML implies some tasks that applications have to perform if they are required to consume or produce information: the task of parsing an XML document, giving specific semantics to the information parsed, and the task of producing an XML document. Our main goal is to create object definitions that can analyze an XML document and automatically create an object definition that can be used abstractly by the application. These objects are able to parse the XML document and gather all the data required to mimic all the information present in the document. This paper introduces xml2pm, a simple tool that can inspect the structure of an XML document and create an object definition (a Perl module) that stores the same information present in the orinial document, but as a runtime object. We also introduce a simple case of how this approach allows the creation of applications based on Web Services in an elegant and simple way.

There are different definitions for ontologies. Different knowledge areas tend to define ontologies in a different way. For computer science, an ontology can be used to describe, in a well defined and structured way, knowledge about a specific domain. These artifacts store rich information that can be reasoned about, this information can also be target of many structured processing functions. There is a diversity of programs that can be implemented to take advantage of these features and produce applications in every area of knowledge. The Ontology Manipulation Language (OML) is a Domain Specific Language (DSL) designed to describe and execute operations that reason about ontologies. These reasoning operations can be used to manipulate and maintain the current information in the ontology, infer new knowledge or concepts, or even produce any kind of side effect. OML is a simple and descriptive language, yet it is powerful enough to implement complex operations or reasoning engines in a clear and efficient way. To actually run programs written in OML a stand alone compiler is available, as well as a mechanism for embedding OML programs in a generic programming language. This allows the quick development of applications that make use of ontologies, by describing ontology related operations in wove OML snippets throughout the code. This mechanism has proven to be a very effective and clear approach for taking advantage of adopting ontologies to represent information, while maintaining the implicit advantages of using a general-goal programming language., This work was partly supported by project CROSS (PTDC/EIACCO/108995/2008), funded by the Portuguese Foundation for Science and Technology.

Today, most developers prefer to store information in databases. But plain filesystems were used for years, and are still used, to store information, commonly in files of heterogeneous formats that are organized in directory trees. This approach is a very flexible and natural way to create hierarchical organized structures of documents. We can devise a formal notation to describe a filesystem tree structure, similar to a grammar, assuming that filenames can be considered terminal symbols, and directory names non-terminal symbols. This specification would allow to derive correct language sentences (combination of terminal symbols) and to associate semantic actions, that can produce arbitrary side effects, to each valid sentence, just as we do in common parser generation tools. These specifications can be used to systematically process files in directory trees, and the final result depends on the semantic actions associated with each production rule. In this paper we revamped an old idea of using a domain specific language to implement these specifications similar to context free grammars. And introduce some examples of applications that can be built using this approach.

Synonyms dictionaries are useful resources for natural language processing. Unfortunately their availability in digital format is limited, as publishing companies do not release their dictionaries in open digital formats. Dicionário-Aberto is an open and free digital synonyms dictionary for the Portuguese language. It is under public domain which makes it usable for any task. Synonyms dictionaries are commonly used for the extraction of relations between words, constructing structures similar to WordNet, or just the extraction of lists of words of specific type. This article presents Dicionário-Aberto, discusses its characteristics and the type of information present on it. Then, we describe an API to help on processing Dicionário-Aberto without the need to tackle with the dictionary format. Finally, we analyze the results on some data extraction experiments.

Languages are born, evolve and, eventually, die. During this evolution their spelling rules (and sometimes the syntactic and semantic ones) change, putting old documents out of use. In Portugal, a pair of political agreements with Brazil forced relevant changes on the way the Portuguese language is written. In this article we will detail these two Orthographic Agreements (one in the thirties and the other more recently, in the nineties), and the challenges present on the automatic migration of old documents spelling to their actual one. We will reveal Bigorna, a toolkit for the classification of language variants, their comparison and the conversion of texts in different language versions. These tools will be explained together with examples of migration issues. As Birgorna relies on a set of conversion rules we will also discuss how to infer conversion rules from a set of documents (texts with different ages). The document concludes with a brief evaluation on the conversion and classification tool results and their relevance in the current Portuguese language scenario. Isbn: 2-9517408-6-7

In our days, the notion, the importance and the significance of parallel corpora is so big that needs no special introduction. Unfortunately, public available parallel corpora is somewhat limited in range. There are big corpora about politics or legislation, about medicine and other specific areas, but we miss corpora for other different areas. Currently there is a huge investment on using the Web as a corpus. This article uncovers GWB, a tool that aims automatic construction of parallel corpora from the web. We defend that it is possible to build high quality terminological corpora in an automatic fashion, just by specifying a sensible Internet domain and using an appropriate set of seed keywords. GWB is a web-spider that works in conjunction with a set of other Open-Source tools, defining a pipeline that includes the documents retrieval from the web, alignment at sentence level and its quality analysis, bilingual dictionaries and terminology extraction and construction of off-line dictionaries.

SIMÕES, Alberto ; CRUZ, Daniela da ; RAMALHO, José Carlos, ed. lit. – “XATA 2010 XML : associated technologies and applications, 8, Vila do Conde, 2010”. [S.l. : s.n.], 2010. ISBN 978-972-99166-9-4., Nowadays XML processing is performed using one of two approaches: using the SAX (Simple API for XML) or using the DOM (Document Ob ject Model). While these two approaches are adequate for most cases there are situations where other approaches can make the solution easier to write, read and, therefore, to maintain. This document presents a rewriting approach for XML documents processing, focusing the tasks of transforming XML documents (into other XML formats or other textual documents) and the task of rewriting other textual formats into XML dialects. These approaches were validated with some case studies, ranging from an XML authoring tool to a dictionary publishing mechanism.

Neste trabalho apresentamos o projecto Procura-PALvras (P-PAL) cujo principal objectivo é desenvolver uma ferramenta electrónica que disponibilize informação sobre índices psicolinguísticos objectivos e subjectivos de palavras do Português Europeu (PE). O P-PAL será disponibilizado gratuitamente à comunidade científica num formato user-friendly a partir de um website a construir para o efeito. Ao utilizar o P-PAL, o investigador poderá fazer uma utilização personalizada do programa ao seleccionar, da ampla variedade de análises oferecidas, os índices que se adequam aos propósitos da sua investigação e numa dupla funcionalidade de utilização: pedir ao programa para analisar listas de palavras previamente constituídas nos índices considerados relevantes para a investigação ou para obter listas de palavras que obedeçam aos parâmetros definidos. O P-PAL assume-se assim como uma ferramenta fundamental à promoção e internacionalização da investigação em Portugal., Fundação para a Ciência e a Tecnologia (FCT), Fundo Europeu de Desenvolvimento Regional (FEDER)

The Marker Hypothesis was first defined by Thomas Green in 1979. It is a psycho-linguistic hypothesis defining that there is a set of words in every language that marks boundaries of phrases in a sentence. While it remains a hypothesis because nobody has proved it, tests have shows that results are comparable to basic shallow parsers with higher efficiency. The chunking algorithm based on the Marker Hypothesis is simple, fast and almost language independent. It depends on a list of closed-class words, that are already available for most languages. This makes it suitable for bilingual chunking (there is not the requirement for separate language shallow parsers). This paper discusses the use of the Marker Hypothesis combined with Probabilistic Translation Dictionaries for example-based machine translation resources extraction from parallel corpora.

Parallel corpora are rich sources of translation resources. This document presents a methodology for the extraction of bilingual nominals (terminology candidates) from parallel corpora, using translation patterns. The patterns proposed in this work specify the order changes that occur during translation and that are intrinsic to the involved languages syntaxes. These patterns are described in a domain specific language named PDL (Pattern Description Language), and are extremely efficient for the detection of nominal phrases., Los corpora paralelos son fuentes ricas en recursos de traducción. Este documento presenta una metodología para la extracción de sintagmas nominales bilingües (candidatos terminológicos) a partir de corpora paralelos, utilizando reglas de traducción. Los modelos propuestos en este trabajo especifican las alteraciones en el orden de las palabras que se producen durante la traducción y que son intrínsecos a la sintaxis de las lenguas implicadas. Estas reglas se describen en un lenguaje de dominio específico llamado PDL (Pattern Description Language) y son sumamente eficientes para la detección de sintagmas nominales., Alberto Simoes has a scholarship from Fundacao para a Computacao Cientifica Nacional and the work reported here has been partially funded by Fundacao para a Ciencia e Tecnologia through project POSI/PLP/43931/2001, co-financed by POSI, and by POSC project POSC/339/1.3/C/NAC.

NAVEGANTE is a generic framework to build superior order proxies for intrusive browsing. This framework provides the means for developing tools that behave as proxies, but perform some processing task on the content that is being browsed. Parallel to this content processing, applications can also run other user-defined functions with different purposes and interfaces, but we'll explain those later. Currently, NAVEGANTE only builds applications that run as CGIs, but this is intended to change in a near future. Applications are built writing programs in NAVEGANTE's Domain Specific Language (DSL). NAVEGANTE is a work in progress. This article aims to describe the current state of development. What applications can be built and how. Also, we identify some implementation problems, and briefly discuss some future improvements. Finally, we try to illustrate most of the concepts described using a couple of case studies.

The creation of a CGI or a WebService as an interface for a command line tool is not as unusual as it may seem. It is extremely usual and useful. There are applications developed as command line tools that can be useful for different purposes, and different kind of users. Some of these users might not be able to run these tools directly. For instance, it is not easy to install a bunch of Perl modules to have a small tool working. For these situations, it is easier to make the tool available in the Web or as a WebService. The problem with making the tool available in these fashions, is that programmers tend to rewrite the tools to incorporate the CGI or XML specific layers. We defend that these CGI or WebService interfaces should use the already available command line tool, without any change. This interface should be able to read a simple textual specification of how the command line tool works, and buid the CGI or XML specific layers automatically. The CGI::Auto module aims this purpose: to encapsulate command line tools in a CGI layer based on a textual specification, transforming the command line tool in a web application.

In this document we describe the use of Text : Rewrite Rules in several tasks related to Speech.

This paper details the SysBio Explorer, a Systems Biology Literature Retrieval and Processing Framework, whose aim relies on the automatic inference of regulatory and metabolic networks based on biomedical literature. The SysBio Explorer does not focus on any organism or problem in particular and encompasses a number of processing and analysis techniques. It works over full-text documents, applying Natural Language Processing techniques and using biomedical dictionaries and ontologies together with hand-made rules. Besides biological entity recognition and relation extraction, document classification, relevance assessment and authoring networks are also within its present scope. The framework is described in terms of its design requirements and implementation decisions, exposing current achievements, but also highlighting present obstacles and future work. Experiments over realworld problems concerning the organisms E. coli, S. cerevisiae and H. pylori are used in its validation., Fundação para a Ciência e a Tecnologia (FCT) - POCI/BIO/60139/2004; POSI/PLP/43931/2001; POSC project POSC/339/1.3/C/NAC., Fundação para a Computação Científica Nacional.

Music Classification is a particular area of Computational Musicology that provides valuable insights about the evolving of compo- sition patterns and assists in catalogue generation. The proposed work detaches from former works by classifying music based on music score in- formation. Text Mining techniques support music score processing while Classification techniques are used in the construction of decision mod- els. Although research is still at its earliest beginnings, the work already provides valuable contributes to symbolic music representation process- ing and subsequent analysis. Score processing involved the counting of ascending and descending chromatic intervals, note duration and meta- information tagging. Analysis involved feature selection and the evalu- ation of several data mining algorithms, ensuring extensibility towards larger repositories or more complex problems. Experiments report the analysis of composition epochs on a subset of the Mutopia project open archive of classical LilyPond-annotated music scores.

This paper describes NATools, a toolkit to process, analyze and extract translation resources from Parallel Corpora. It includes tools like a sentence-aligner, a probabilistic translation dictionaries extractor, word-aligner, a corpus server, a set of tools to query corpora and dictionaries, as well as a set of tools to extract bilingual resources., Alberto Simoes has a scholarship from Fundacao para a Computacao Cientifica Nacional and the work reported here has been partially funded by Fundacao para a Ciencia e Tecnologia through project POSI/PLP/43931/2001, co-financed by POSI, and by POSC project POSC/339/1.3//NAC.

As ferramentas de tradução assistida por computador tentam reutilizar as traduções realizadas pelo tradutor sempre que uma frase semelhante tenha sido já traduzida. Para o intercâmbio destes documentos foi definido um formato denominado TMX (Translation Memory Exchange) baseado em XML. Este tipo de documento ganha facilmente tamanhos incomportáveis para o seu processamento com métodos tradicionais. Neste artigo propomos uma metodologia de ordem superior para o processamento de documentos de estrutura repetitiva (em que se inserem as memórias de tradução) com uma abordagem baseada na conjunção de SAX e DOM. São apresentados vários exemplos de filtros sobre memórias de tradução bem como um conjunto de medidas da sua eficiência., Este trabalho foi parcialmente financiado pela Fundação para a Ciência e Tecnologia de Portugal pela bolsa POSI/PLP/43931/2001, e co-financiado pelo POSI, dentro da Linguateca.